A growing number of college applications has presented an annual challenge for college admissions in the US. In response to this challenge, admission offices have often relied on standardized test scores to parse their large applicant pools into viable subsets. However, this approach may be subject to bias in test scores and fails to work in test-optional admissions. In this work, we explore a machine learning-based approach to replace the role of standardized tests in subset generation while taking into account a wide range of factors extracted from student applications to support a more holistic review. We evaluate the approach on data from an undergraduate admissions office at a selective US institution and discuss how machine learning can be leveraged to support human decision-making in college admissions.